Geomorphology 102 (2008) 615–623
Contents lists available at ScienceDirect
Geomorphology
j o u r n a l h o m e p a g e : w w w. e l s e v i e r. c o m / l o c a t e / g e o m o r p h
Post-glacial fluvial response and landform development in the upper Muskegon River
valley in North-Central Lower Michigan, U.S.A.
Alan F. Arbogast a,⁎, Juleigh R. Bookout a, Bradley R. Schrotenboer a, Amy Lansdale b,
Ginny L. Rust c, Victorino A. Bato d
a
Department of Geography, 123 Geography, Michigan State University, East Lansing, MI, 48824-1115, USA
ENVIRON International Corp, 4350 North Fairfax Drive, Suite 300, Arlington, VA 22203, USA
Department of Earth and Planetary Sciences, University of New Mexico, Albuquerque, New Mexico 87131, USA
d
National Mapping and Resource Information Authority of the Philippines, Makati City, Philippines
b
c
A R T I C L E
I N F O
Article history:
Received 29 November 2007
Received in revised form 10 June 2008
Accepted 11 June 2008
Available online 20 June 2008
Keywords:
Muskegon River
Michigan
Stream terraces
Paleomeanders
A B S T R A C T
This study focuses on the upper part of the Muskegon River system in north-central Lower Michigan and is
the first to reconstruct the post-glacial history of fluvial landform development in the core of North America's
Great Lakes region. Results indicate that the upper Muskegon River valley contains four alluvial terraces and
numerous paleomeanders. Radiocarbon dating of peats within these old channels provides a good
chronology for stream behavior and landform development. The T-4 terrace is a paired Pleistocene
outwash/lacustrine surface that probably formed about 12,500 years ago. The T-3 terrace is a fill-strath
surface that was cut between about 12,000 and perhaps 9500 years ago. The geometry of macromeanders on
this surface suggests that stream discharge was ∼ 8 times greater than during the Holocene.
The Pleistocene/Holocene transition is marked by a major period of downcutting that likely began as the
climate warmed/dried and sediment yield diminished. This period of downcutting potentially lasted through
the drier middle Holocene, creating a 6-m-high escarpment in the valley. The Muskegon River then began to
aggrade when the climate became wetter. Subsequently the river again incised, creating the paired T-2
terrace, about 3400 years ago when the climate became still wetter. T-2 paleomeanders indicate that stream
discharge at this time was consistent with the modern river. In the past 2500 years, the stream has
constructed a poorly defined complex of T-1 terraces. These surfaces likely formed due to complex response
associated with more variable climate. This study demonstrates that the upper Muskegon River has a similar
post-glacial history as streams on deglacial and periglacial landscapes elsewhere in the world.
© 2008 Elsevier B.V. All rights reserved.
1. Introduction
The importance of fluvial landforms in Quaternary landscape
studies has long been recognized (e.g., Davis, 1909; Penck and
Bruckner, 1909; Fisk, 1944; Dury, 1970; Knox, 1975; Harvey et al.,
1981; Ballantyne and Whittington, 1999; Howard et al., 2004). In the
context of these investigations, internal adjustments and landform
development have been linked to a number of variables, including
base level change (e.g., Davis, 1909; Chorley, 1963; Bull, 1990), climate
fluctuations (e.g., Knox, 1984; Van Nest and Bettis, 1990; Arbogast and
Johnson, 1994; Hsieh and Knuepfer, 2001; Bridgland et al., 2004), and
internal complex responses (e.g., Schumm and Parker, 1973; Womack
and Schumm, 1977; Minako and Tetsuji, 2006). Taken together, these
studies have demonstrated that stream systems are excellent indirect
indicators of environmental change.
⁎ Corresponding author. Tel.: +1 517 355 5262; fax: +1 517 432 1671.
E-mail address: [email protected] (A.F. Arbogast).
0169-555X/$ – see front matter © 2008 Elsevier B.V. All rights reserved.
doi:10.1016/j.geomorph.2008.06.008
One particular focus of stream studies has been the response of
fluvial systems in glaciated and periglacial areas at the end of the most
recent ice age (marine Oxygen Isotope Stage 2) and the subsequent
evolution of landforms during the Holocene. It is well documented, for
example, that braided meltwater streams around the world developed
meandering channels when sediment yield decreased as deglaciation
progressed (Schumm and Brakenridge, 1987; Knox, 1995; Starkel,
1995; Blum and Törnqvist, 2000). Similar changes have been reported
for streams in periglacial areas (e.g., Kozarski, 1991; Kasse et al., 1995;
Straffin et al., 2000).
Other streams, such as some in the western Canadian prairies (e.g.,
Bryan et al., 1987) and Europe (Schumanski, 1983; Starkel et al., 1996;
Sidorchuk et al., 2001), changed from systems with macromeanders
during the late Pleistocene to those with much smaller Holocene
meandering patterns. Another observation is that many streams
began downcutting at the Pleistocene/Holocene transition because of
reduced sediment load, resulting in distinct, paired terraces. This type
of channel adjustment has been reported in Europe (Starkel, 1991;
Walker, 1995; Tebbens et al., 1999; Howard et al., 2004), the western
616
A.F. Arbogast et al. / Geomorphology 102 (2008) 615–623
Canadian prairies (Rains and Welch, 1988; Campbell and Campbell,
1997; Otelelaar, 2002), and the state of New York in the eastern U.S.
(Scully and Arnold, 1981).
Following the adjustments at the Pleistocene/Holocene transition,
many streams in formerly glaciated and periglacial areas subsequently
underwent a series of changes in the Holocene that led to the creation
of terraces. In northern Britain, for example, flights of terraces
developed during the middle and late Holocene due to climate
variation and human activity (Howard et al., 1999). Along the upper
Susquehanna River in the state of New York, 2 terraces and the
floodplain have formed since deglaciation due to climate changes. A
series of alluvial terraces has also been recognized in Alberta, Canada
that formed since the ice retreated (Otelelaar, 2002; Evans et al.,
2004).
Although much is known about the post-glacial landform development of river systems in many deglacial and periglacial areas, little
is known about such histories in the core of the Great Lakes region in
North America. Abundant fluvial research has been conducted on
streams in the state of Wisconsin, but has largely focused on Holocene
alluviation processes (e.g., Knox, 1972), the Holocene flood history of
the upper Mississippi valley (Knox, 1993, 2006), and terrace formation
due to European settlement (Woltemade, 1994). A limited amount of
fluvial research has also been conducted in Michigan, but has focused
primarily on stream response near the Great Lake's coasts during the
Nipissing transgression in the middle Holocene and subsequent
cutting and filling cycles in the late Holocene (Monoghan and Lovis,
2005).
The primary goal of this study is to reconstruct the chronology of
stream behavior and landform development in the Great Lakes region
and relate it to patterns observed elsewhere in the world (e.g., Scully
and Arnold, 1981; Schumanski, 1983; Bryan et al., 1987; Sidorchuk
et al., 2001; Howard et al., 2004). It focuses on the upper Muskegon
River valley in north-central Michigan where a variety of stream
terraces and related landforms have been identified that spans the
post-glacial history of the system.
2. Regional setting
The Muskegon River is a meandering stream that originates in
north-central Lower Michigan at Houghton Lake (Fig. 1). From this
locality, the river flows about 340 km before it enters Lake Michigan at
the city of Muskegon. Over its course, the stream drops 189 m and
drains about 6000 km2 of area (O'Neal, 1997). The study area for this
project focuses on the upper half of the Muskegon River, extending
∼150 km from Reedsburg Dam near Houghton Lake southwest to the
US-10 bridge in the town of Evart (Fig. 1). Along this length, the river
descends 45 m, winding its way through numerous meanders, split
channels, and oxbow lakes. The average gradient of the Muskegon
River in this reach of the system is ∼ 0.005. Mean discharge of the
Muskegon River at Evart ranges from a low of about 14 m3/s in late
summer and early fall to a high of about 60 m3/s at the height of the
spring snow melt (USGS, 2004).
Prior to the initial development of the Muskegon River, northern
Michigan was covered by glacial ice for much of the Late Wisconsin
glacial period (e.g., Eschman, 1985; Blewett and Winters, 1995;
Schaetzl and Weisenborn, 2004). Final deglaciation of the study area
occurred ∼ 16 ka when ice retreated from southern Michigan north to
the Straits of Mackinac and to the west and east toward the Lake
Michigan and Lake Huron basins, respectively (e.g., Schaetzl and
Weisenborn, 2004). Although details are vague, Tardy (2005) believes
that a large outwash plain (known locally as the Saint Helen outwash
plain) was constructed in the area where the Muskegon River now
originates during this deglacial phase. Soil evidence (e.g., laminae,
thick clay lenses) suggests that much of the outwash plain was
Fig. 1. The location of the Muskegon River in Lower Michigan. Inset maps show the position of Michigan in the U.S and the extent of the study area from Reedsburg Dam to Evart.
A.F. Arbogast et al. / Geomorphology 102 (2008) 615–623
subsequently inundated by a meltwater lake or lakes (Tardy, 2005) as
the ice receded further. Houghton Lake is probably a remnant of this
late Pleistocene lake system.
In addition to the outwash and lacustrine sediments in the region,
a portion of the study area consists of complex mix of swamps, bogs,
and sand dunes. A large bog occurs, for example, northwest of the
uppermost reaches of the Muskegon River west of Houghton Lake. An
array of other bogs are scattered throughout the area. A small
(∼ 20 km2) dune field occurs on the east side of the Muskegon River
about 25 km south of Houghton Lake.
The climate of the Muskegon watershed is classified as humid
continental, with cold winters and warm, humid summers. Most
precipitation occurs during the late summer months as moisture from
the Gulf of Mexico flows north into the area. Winter precipitation falls
as snow, with the heaviest snowfalls brought in by fronts moving east
across Lake Michigan. Some additional snowfall accumulates from
lake-effect precipitation, especially to the southwest. Total precipitation generally increases from NE to SW across the watershed, ranging
from an average of 71 cm near Houghton Lake to 94 cm around
Newaygo (Eichenlaub et al., 1990). Vegetation within the study area is
largely mixed conifer–hardwood forests composed of White pine
(Pinus strobes), Trembling aspen (Populus tremuloides), oak (e.g.,
Quercus rubra), and Paper birch (Betula papyrifera). Maple (e.g., Acer
rubrum), American beech (Fagus grandifolia), and Eastern hemlock
(Tsuga canadensis) are also found in smaller numbers (Bernabo, 1981).
617
3. Materials and methods
The first step we took in this study was to analyze the study reach
by investigating county soil surveys, aerial photographs, and 1:24,000scale topographic maps. Through this analysis, we identified three
prominent landforms: i) an irregular valley wall that essentially
bounds the study area; ii) a broad, paired terrace; and iii) a
considerable terrace escarpment on either side of the present river
channel that confines the river to an “inner valley.” Within this inner
valley, several poorly defined terraces appeared to be also present. In
order to verify these observations, we entered the field with the intent
to characterize and map all of the terraces along the river. However,
poor road access and dense vegetation in most places precluded our
ability to construct a detailed map and, therefore, we adjusted the
scope of our project to instead focus on mapping the valley wall and
prominent terrace escarpment near the stream. We also sought to
characterize the valley in cross-section and determine the ages of the
more well-defined terraces where we could distinguish between them.
In an effort to identify the sequence of terraces in the valley, we
established six cross-sectional transects that were surveyed with a
Trimble Geo Explorer III GPS unit. These transects were evenly
distributed across space between the headwaters and Evart and
extended between the uplands on either side of the valley. Points were
collected at 50-m intervals on horizontal surfaces and at targeted
locations such as the top and base of escarpments. The data were
Fig. 2. Map of study sites in the upper part of the Muskegon River valley.
618
A.F. Arbogast et al. / Geomorphology 102 (2008) 615–623
Table 1
Radiocarbon ages presented in this study. All ages are derived from basal peats in paleomeanders
Site
Terrace
Depth (m)
Lab #
Conventional age (2σ)
Calibrated agea (2σ)
13 b
1
2
3
4
5
5
5
6
6
6
7
8
9
10
11
12
13
14
15
T-1
T-3
T-3
T-1
T-3
T-3
T-3
T-2
T-2
T-2
T-3
T-2
T-1
T-2
T-2
T-1
T-3
T-2
T-1
.75
1.5
.88
1.52
1.50
1.45
1.61
0.15
0.22
0.60
1.8
1.52
0.46
1.1
1.1
0.91
2.45
2.45
0.75
Curl-5510
Curl-5511
Curl-5512
Beta-161766
Curl-5514
Beta-172561
Beta-172560
Beta-161771
Beta-161770
Beta-161769
Beta-179270
Beta-161772
Beta-161768
Beta-179269
Beta-198066
Beta-198065
Beta-198064
Beta-198063
Beta-198062
2150 ± 80
8700 ± 100
3590 ± 80
1800 ± 80
9540 ± 110
9890 ± 80
9930 ± 80
Modern
3790 ± 80
1210 ± 80
9080 ± 100
3350 ± 80
1190 ± 80
4160 ± 60
5500 ± 80
520 ± 80
10320 ± 140
3590 ± 100
1200 ± 80
2335–1986
10,148–9500
4091–3688
1830–1610
11,169–10,575
11,300–11,200
11,530–11,220
Modern
4280–4010
1250–1050
10,260–10,180
3680–3470
1230–990
4830–4570
6330–6210
630–510
12360–11,640
4060–3720
1240–1000
−25.5
−26.5
–25.1
−25.9
−26.0
− 27.8
− 27.3
−26.6
− 27.1
−26.9
− 27.6
−25.1
−28.3
−24.5
−28.0
−25.5
−31.1
− 27.3
−24.5
C
a
Calibrated from conventional δ13C-corrected radiocarbon age to calendar years using a tree-ring curve. All calibrations reported here were based upon the 20-yr atmospheric
curve (e.g., Linick et al., 1985; Stuiver et al., 1986). Program used is discussed in Stuiver et al. (1998).
b
For a discussion of the δ13C-correction procedure, see Stuiver and Polach (1977) and Taylor (1987).
differentially corrected, resulting in horizontal and vertical accuracies
of ∼ .6 m and 1.5 m, respectively. The overall goal in this phase of the
project was to learn enough about the valley to create a representative
cross-section that illustrates the various terraces.
In addition to mapping and characterizing cross-sections of the
river valley, another goal of the project was to determine the age of the
terraces. Although no dateable material was present in any of the
cutbanks investigated in the area, we discovered that several
prominent meander scars contained thick (N1 m) deposits of peat.
Assuming that peat accumulation began shortly after oxbow formation, the age of basal organics in meander scars represents a
reasonable minimum-limiting estimate for the associated stream
terrace. This sampling method has been successfully used to estimate
the minimum-limiting age of beach ridges immediately lakeward of
peat-filled swales along the Great Lakes (Thompson, 1992; Thompson
and Baedke, 1997; Baedke and Thompson, 2000). Leigh (2006) used a
similar strategy to date paleomeanders in streams on the Atlantic
Coastal Plain in the southeastern U.S.
We sampled the peat at 15 sites (Fig. 2) by hand-augering to a
depth a few centimeters below the contact with underlying sands. The
resultant core at the peat–sediment interface was then split open and
an organic sample was collected immediately above the contact with
the underlying sand from the interior of the core to avoid contamination. The samples were then wrapped in aluminum foil and frozen
until they were shipped for age determination via AMS dating. Fifteen
samples were processed at Beta Analytic in Miami, FL, and four were
analyzed at the INSTAAR Laboratory in Boulder, CO. In order to correct
for long-term variations in the radiocarbon timescale, all dates were
calibrated to the tree-ring curve established by Stuiver et al. (1998),
allowing radiocarbon dates to be adjusted to calendar years before
present (i.e., cal. YBP).
4. Results
4.1. Stream terraces and radiocarbon dates
Results from this study indicate that four stream terraces are
present in the upper Muskegon River valley. Radiocarbon dates from
these surfaces range in age from the late Pleistocene to the late
Holocene (Table 1). The geomorphic relationship of these surfaces can
be seen in the schematic transect presented in Fig. 3. In addition, the
stratigraphy of sites and related terraces is presented in Fig. 4.
The T-4 and T-3 terraces collectively form the dominant alluvial
surfaces in the valley and are mapped as part of the local outwash
plain by Tardy (2005). Field examination of the underlying fill of
cutbank exposures of the T-4 and T-3 terraces supports this hypothesis
as it consists largely of sand with lenses of gravel. The T-4 surface is the
most extensive of these landforms, extending to the east for more than
Fig. 3. Schematic cross-section of landforms in the upper part of the Muskegon River valley. The larger-scale map showing detail of the inner valley is magnified 3×.
A.F. Arbogast et al. / Geomorphology 102 (2008) 615–623
619
Fig. 4. Stratigraphy and radiocarbon dates from sites investigated in this study.
1 km in places and up to 0.5 km on the west side of the stream. The T-3
terrace lies between ∼ 1.5 and 2.0 m below the T-4 surface and is lessextensive, ranging from about 100 to 500 m wide.
Evidence indicates that the T-3 surface was created by a meandering ancestral Muskegon River that planed off the upper portion of
T-4 fill, making the T-3 a fill-strath terrace. This evidence can be seen
in Fig. 5 in the form of 3 large paleomeanders in the upper part of the
study area. Each of these channels is ∼ 90 m wide and has a radius of
curvature of ∼325 m. The meander wavelength is ∼800 m. Peat
thickness in these macromeanders ranges from ∼2.5 m at sites 5, 13,
and 14 to 0.9 m at site 3 (Fig. 4).
Calibrated radiocarbon dates from basal peat in the macromeanders provide minimum-limiting estimates for the T-3 terrace. Of the
six age estimates derived from this surface, 5 of them range from
12,360 to 9500 cal. YBP. At Site 5 two lenses of sand were observed
with interbedded peats (Fig. 4). Ages derived from these peats all fell
between about 11,500 and 10,500 cal. YBP. At another site (Site 3) an
age of 4091–3688 cal. YBP was derived from basal peat on the T-3
terrace. Overall, the older dates from the T-3 terrace indicate that the
ancestral Muskegon River occupied this surface in the latest
Pleistocene. The younger date is a bit problematic and could be
explained by the following: i) the sample was contaminated with
younger organic material, resulting in anonymously younger ages; ii)
peat did not begin to form in the paleomeander for thousands of years
after they were abandoned; and iii) the Muskegon occupied the T-3
surface from the late Pleistocene into the middle Holocene. This latter
hypothesis is rejected for reasons outlined below.
Although the precise chronology of T-3 surface occupation by the
Muskegon River is unknown, it clearly became a terrace following a
major period of downcutting in the valley. This period of entrenchment produced a distinct but narrow inner valley throughout the
study area that is bordered on both sides of the river by a ∼6-m-high
escarpment that separates the T-3 terrace from younger surfaces
(Fig. 3). Two terraces were identified in this inner valley, a T-2 that lies
directly below the T-3, and a T-1. The escarpment between the T-2 and
T-1 is ∼ 1 m high and is clearly visible only in scattered forest clearings
and along the river where both surfaces are present. The T-1 terrace
actually appears to be a complex of poorly defined terraces that differ
slightly in elevation on a local level.
One of the most notable features of the T-2 and T-1 surfaces is that
they both contain numerous paleomeanders and oxbow lakes (e.g.,
Fig. 5). In contrast to the macromeanders on the T-3 terrace, the
abandoned channels on the lower surfaces are substantially smaller.
The width of these old channels is ∼ 30 m, the radius of curvature
ranges from ∼50 to 75 m, and the meander wavelength is from ∼ 200
to 350 m. These geometric aspects are very consistent with the
modern river.
A major goal of the study was to estimate when the extensive
period of channel entrenchment occurred that created the T-3 terrace
and to determine the age of the T-2 and T-1 surfaces. In that context,
12 radiocarbon ages were obtained from peats in paleomeanders
within the inner valley, with seven obtained from the T-2 and five
from the T-1. The ages from the T-2 surface range from 6330 to
3720 cal. YBP, with two ages falling between 4830 and 4010 cal. YBP
(Table 1; Fig. 4). Ages derived from the T-2 at site 6 were
stratigraphically inverted. The basal sample (0.60 cm deep) at this
Fig. 5. Aerial photograph showing the nature and ages of late Pleistocene and Holocene
paleomeanders in the upper part of the Muskegon River valley. Note the difference in
size between the paleomeanders of differing ages.
620
A.F. Arbogast et al. / Geomorphology 102 (2008) 615–623
site provided an age of 1250–1050, whereas a sample from
0.22 cm yielded an estimate of 4280–4010. We conclude that the
peat at this site was somehow disturbed, perhaps by tree throw
or burrowing animals. Ages obtained from the T-1 terrace range
from 2335 to about 500 cal. YBP. Of the five ages derived from this
surface, four occur within a range that extends from 1830 to 1000 cal.
YBP.
5. Discussion
This study reconstructed the chronology of post-glacial stream
processes and associated landform development in the upper part of
Michigan's Muskegon River system. Results indicate that four stream
terraces occur along this reach. The development of these surfaces
appears to be related to distinct periods of environmental change
and associated stream responses (Table 2). The initial history of the
system is uncertain and is probably related to the deglaciation of the
area. The earliest ancestral Muskegon River may have been
associated with the braided stream system that formed the extensive
Saint Helen outwash plain when ice retreated from the area around
16 ka. Alternatively, the stream originated, or was somehow
modified, when the meltwater lake(s) that apparently existed in
the area (Tardy, 2005) for an unknown period of time began to drain.
Although the earliest history of the Muskegon River is not
presently known, evidence indicates that by the latest Pleistocene
the stream had downcut ∼ 2 m to form the T-4 terrace. This surface is a
paired terrace that is underlain by the uppermost outwash/lacustrine
sediments. During this time the ancestral Muskegon River had a large
meandering channel that cut an erosional surface into outwash and
lacustrine sediments where they are present. This cut surface
ultimately became the T-3 terrace following a subsequent period of
downcutting.
The oldest radiocarbon date obtained from peat contained within a
macromeander on the T-3 terrace is 12,360–11,640 cal. YBP. This date
provides a minimum-limiting estimate for the Muskegon's initial
downcutting event and suggests that the river had developed into a
meandering stream by perhaps 13–12.5 ka. This period of time marks
the initial expansion of spruce into the region (Webb et al., 1983),
which may have stabilized channel banks and reduced the yield of
sandy bank material such that a change in channel form from braided
to meandering occurred (e.g., Carson, 1984; Bridge, 2003; Leigh,
2006). It is also possible that the change occurred in part because
sediment supply from periglacial sources to the north diminished.
Table 2
Chronology of stream behavior and environmental change in upper Muskegin River
valley
Years
ago
0
1000
2000
3000
4000
5000
6000
7000
8000
9000
10,000
11,000
12,000
13,000
14,000
15,000
16,000
Stream response
Environment
Aggradation/incision; complex response
Aggradation; river is meandering
More variable climate; Northern
hardwood forest
Cooler/more moist;
pine-dominated forest
Cool/moist; pine-dominated forest
Incision
Warmer/drier, oak forest
Macromeanders; Q = ∼8 × Holocene;
Planing of T-3 terrace
Cool/moist; Spruce forest
Uncertain; draining of lake? Slight
incision
Deglaciation
Cool/dry; Spruce-sedge parkland
Incision
Cold/dry; Tundra
Dashed lines between stages reflect dating uncertainties.
This kind of relationship as has been reported for streams in Europe
(Kozarski, 1991; Kasse et al., 1995; Straffin et al., 2000).
Although the reason for the change in channel pattern is unknown
with certainty, it is clear from the size of the macromeanders on the T3 terrace that the ancestral Muskegon River had a much greater
discharge than the stream had during the Holocene or today. These
macromeanders are generally consistent in size and form with those
reported along the glacial margin in Europe (Schumanski, 1983;
Starkel et al., 1996; Sidorchuk et al., 2001) and Alberta, Canada (Bryan
et al., 1987). Macromeanders of very similar size (i.e., channel width,
meander wavelength, radius of curvature) as those observed in this
study have also been reported along several streams on the Atlantic
Coastal Plain in the southeastern U.S. Leigh and Feeney (1995)
estimated that mean annual bankfull discharge in these ancient
streams was between ∼ 400 and 500 m3/s, whereas modern average
bankfull discharge is between ∼60 and 85 m3/s. According to Leigh
and Feeney (1995), this higher discharge occurred because the climate
between ∼31–28 ka and ∼8.5–4.5 ka (when the meanders formed)
was about 10%–30% wetter than the current one.
Given the general geometric consistency of macromeanders
reported in this study with those in the southeastern U.S. (Leigh and
Feeney, 1995), we believe that mean annual bankfull discharge of the
ancestral Muskegon River during the late Pleistocene was at least
∼400 m3/s. This amount of flow is ∼8 times greater than average
modern spring discharge in the river and probably occurred because i)
the meltwater lake(s) in the area (Tardy, 2005) were draining
significantly, ii) the regional water table was very high due to the
recently melted glacier in the area (e.g., Webb et al., 2004), or iii) a
combination of the two.
Radiocarbon dates from macromeanders on the T-3 terrace suggest
that the ancestral Muskegon River maintained this high discharge for
the last ∼2500 to 2000 years of the Pleistocene and perhaps into the
very early Holocene. This conclusion is based on the fact that the
majority of minimum-limiting age estimates from the macromeanders range from between ∼12,360 and 9500 cal. YBP. This period of
time nicely correlates to the interval during which spruce forests
dominated similar low-lying and moist outwash areas in the upper
Midwest (e.g., Pregitzer et al., 2000). The Muskegon River was
apparently in static equilibrium (Bull, 1991) during this phase of its
history as it migrated laterally and planed the fill-strath surface that
became the T-3 terrace.
This period of equilibrium ended when a major period of incision
began in the upper Muskegon valley, one that elevated the T-3 terrace
relative to the active flood plain and created a 6-m-high escarpment.
Circumstantial evidence suggests that this period of downcutting
occurred at the Pleistocene/Holocene transition when the climate
began to warm and stream discharge decreased. This form of channel
response at the Pleistocene/Holocene transition has been documented
in the glacial and periglacial parts of Europe (e.g., Starkel, 1991;
Walker, 1995; Tebbens et al., 1999; Howard et al., 2004), along the Bow
River in Alberta, Canada (Otelelaar, 2002), and along the upper
Susquehanna River in New York (Scully and Arnold, 1981). A variety of
evidence (e.g., Webb et al., 1983; Krishnamurthy et al., 1995; Bartlein
et al., 1998; Kutzbach et al., 1998; Davis et al., 2000; Booth et al., 2002)
indicate that this climate transition occurred at about 10 ka in the
Great Lakes region. In Lower Michigan, this environmental transition
resulted in spruce forests being replaced by oak-dominated forests
(Manny et al., 1978).
Although we do not know with certainty, it appears that early
Holocene entrenchment extended through the middle Holocene dry
period (a.k.a. Hypsithermal/Altithermal) that is well documented in
the region (Webb et al., 1983; Krishnamurthy et al., 1995; Bartlein
et al., 1998; Kutzbach et al., 1998; Davis et al., 2000; Booth et al.,
2002, 2004). According to Davis et al. (2000), for example, the
middle Holocene climate in northern Michigan was about 9% drier
than the modern one. Knox (1972) argued that the overall more arid
A.F. Arbogast et al. / Geomorphology 102 (2008) 615–623
middle Holocene climate in the upper Midwest paradoxically
resulted in more intense storms and flooding on a periodic basis.
In the central U.S., this instability is linked to valley alluviation due to
reduced vegetation on hillslopes and higher yield of fine-grained
sediment (Knox, 1972; May, 1989; Martin, 1992; Van Nest and Bettis,
1990; Arbogast and Johnson, 1994; Baker et al., 2000). Such a
reduction in vegetation cover likely did not happen in central Lower
Michigan, however, as the late Pleistocene spruce forest merely
changed to one dominated by pine and oak (Webb et al., 1983). As a
result, the Muskegon River may have continued downcutting when
periodic high discharges swept through the system in the middle
Holocene.
Following the period of channel entrenchment that created the
high escarpment below the T-3 terrace, the Muskegon River switched
to an aggradational system. During this period of alluviation, the river
deposited fine-textured sediment that partially filled the inner valley
bordered by the high escarpment. This fill now underlies the T-2
terrace, which is preserved as a paired surface in the study area.
Numerous paleomeanders occur on this surface, indicating that the
Muskegon was a stream in dynamic equilibrium with a low flow
velocity (Knighton, 1984). The oldest radiocarbon date obtained on
basal peats from a T-2 paleomeander was 6330–6210 cal. yrs B.P.,
indicating that post-entrenchment alluviation began sometime before
that time. According to several studies (e.g., Davis et al., 2000; Jackson
and Booth, 2002; Booth et al., 2004), the climate of northern Michigan
became more mesic about 6 ka. Given the constraints associated with
assigning precise time periods through radiocarbon dating, we
propose that alluviation within the Muskegon River system is
generally associated with the return to a more humid climate between
about 7 and 6 ka and the related delivery of more sediment
consistently to the stream channel.
Following the period of alluviation that partially filled the inner
Muskegon valley, the river apparently occupied the T-2 surface for a
distinct period of time. Supporting evidence for this conclusion is that
numerous paleomeanders occur on the T-2 terrace. These paleomeanders are significantly smaller with respect to their width, radius of
curvature, and meander wavelength than the macromeanders on the
T-3 terrace. Instead, they are consistent in size with the modern
stream, suggesting that the average Holocene stream discharge was
approximately 30–40 m3/s. This kind of change to smaller Holocene
channel geometry has been reported in Europe (Schumanski, 1983;
Starkel et al., 1996; Sidorchuk et al., 2001), the western Canadian
prairies (Bryan et al., 1987) and the Atlantic Coastal Plain in the
southeastern U.S. (Leigh, 2006). Radiocarbon ages from basal peats in
several of the paleomeanders along the upper Muskegon River range
from 6330 to 3720 cal. YBP, suggesting that the stream flowed on this
surface for ∼ 2500 years.
Paleoclimate evidence from Michigan indicate that ∼ 3200 cal. YBP
the climate of the area became cooler and more moist (Bernabo, 1981;
Jackson and Booth, 2002; Booth et al., 2004), with pine forest
dominating the landscape. In the upper Muskegon River valley, this
climate shift appears to be associated with a period of minor
downcutting that formed the T-2 terrace. According to Bull (1991),
this kind of stream degradation can be attributed to an increase in the
total annual stream power, which would have resulted from a wetter
climate. This type of climate shift/response has been linked with an
episode of downcutting in the western Canadian prairies (Campbell
and Campbell, 1997). The length of this period of limited entrenchment along the upper Muskegon River is uncertain, but may have been
for several hundred years.
Following this latter period of downcutting, the Muskegon River
began to respond in a complex manner that formed a series of illdefined and unpaired surfaces close to the modern flood plain that are
lumped into the T-1 terrace complex. The stream continued to
meander during this phase, resulting in numerous paleochannels and
oxbow lakes, The first stage of this chronology began with a period of
621
valley filling that again deposited fine-grained sediment to a point
about 1 m below the T-2 terrace. A radiocarbon date of 2335–1986 cal.
yrs B.P. was obtained from a paleochannel at this elevation, suggesting
that valley filling may have began approximately 2500 years ago. Since
that time, the stream has undergone a series of subtle cutting and
filling cycles, resulting in the T-1 terrace complex. Radiocarbon ages
from paleomeanders in this complex of minor terraces range from
∼2300 cal. YBP. to the present.
The apparent onset of the more complex late Holocene stream
history correlates reasonably well with the paleoclimate record from
northern Michigan. According to several studies (Jackson and Booth,
2002; Booth et al., 2004; Webb et al., 2004), the late Holocene climate
of the region has fluctuated more than earlier periods, with
alternating wet and dry intervals. Vegetation during this period of
time has consisted of northern hardwoods. We propose that this more
variable climate has provoked a complex response (Schumm and
Parker, 1973; Womack and Schumm, 1977; Bull, 1991) in the system for
the past 2500 years.
6. Conclusion
This study reconstructed the post-glacial chronology of stream
behavior and landform development along the upper Muskegon River
in north-central Lower Michigan. As a result, it is the first investigation
to compare the history of a stream in the core of the Great Lakes region
in North America with other streams around the world in deglacial
and periglacial areas. This chronology is based on radiocarbon dates
obtained from basal peats in paleomeanders on stream terraces.
Results indicate that the upper valley contains four alluvial terraces
and that stream discharge varied significantly between the late
Pleistocene and Holocene. The T-4 terrace is a Pleistocene outwash/
lacustrine surface that probably formed during a period of downcutting about 12,500 years ago. The T-3 terrace is a fill-strath surface
that was cut between about 12,000 and perhaps 9500 years ago.
During this time the discharge of the river was perhaps 8 times greater
than the modern stream, resulting in the formation of macromeanders
consistent in size and form with late Pleistocene rivers in glacial and
periglacial areas elsewhere in the world.
Following this period of dynamic equilibrium, a major period of
downcutting likely began at the Pleistocene/Holocene climate transition, one that also occurred in streams on similar landscapes
elsewhere in the world. In north-central Lower Michigan, this period
of downcutting was likely triggered because stream discharge
decreased as the climate became warmer and drier. We believe that
this period of downcutting lasted through the drier part of the middle
Holocene, creating a 6-m-high escarpment along both sides of the
valley.
Following the period of middle Holocene entrenchment, the
Muskegon River began to aggrade at the onset of a wetter climate
period. This period of aggradation may have lasted about 2500 years.
The size of paleomeanders from this time period indicates that the
river was similar in size with the modern river and that discharge had
decreased significantly from the late Pleistocene. Subsequently, the
stream apparently incised again about 3400 years ago when the
climate became still wetter. This period of incision created the paired
T-2 terrace. In the past 2500 years, the stream has constructed a poorly
defined complex of T-1 terraces. These surfaces may have evolved
because the regional climate in the late Holocene has been more
variable than at other times.
Acknowledgements
We would like to acknowledge Cathryn Dowd for her early work
with this project. In addition, we would like to thank Eric Manschott
for his assistance in the field. Lastly, this project was supported in large
part by NSF grant #BCS-0117612. Comments from two anonymous
622
A.F. Arbogast et al. / Geomorphology 102 (2008) 615–623
reviewers substantially improved the manuscript and are much
appreciated.
References
Arbogast, A.F., Johnson, W.C., 1994. Climatic implications of the late Quaternary alluvial
record of a small drainage basin in the central Great Plains. Quat. Res. 41, 298–305.
Baedke, S.J., Thompson, T.A., 2000. A 4,700-year record of lake level and isostasy for
Lake Michigan. J. Great Lakes Res. 26, 416–426.
Baker, R.G., Bettis III, E.A., Fredlund, G.G., Mandel, R.D., 2000. Holocene environments of
the central Great Plains: multi-proxy evidence from alluvial sequences, southeastern Nebraska. Quat. Int. 67, 75–88.
Ballantyne, C.K., Whittington, G., 1999. Late Holocene floodplain incision and alluvial
fan formation in the central Grampian Highlands, Scotland: chronology, environment, and implications. J. Quat. Sci. 14, 651–671.
Bartlein, P.J., Anderson, K.H., Anderson, P.M., Edwards, M.E., Mock, C.J., Thompson, R.S.,
Webb, R.S., Webb III, T., Whitlock, C., 1998. Paleoclimate simulations for North
America over the past 21,000 years: features of the simulated climate and
comparisons with paleoenvironmental data. Quat. Sci. Rev. 17, 549–585.
Bernabo, J.C., 1981. Quantitative estimates of temperature changes over the last
2,700 years in Michigan based on pollen data. Quat. Res. 15, 143–159.
Blewett, W.L., Winters, H.A., 1995. The importance of glaciofluvial features within
Michigan's Port Huron moraine. Ann. Assoc. Am. Geogr. 85, 306–319.
Blum, M.D., Törnqvist, T.E., 2000. Fluvial responses to climate and sea-level change: a
review and look forward. Sedimentology 47, 2–48.
Booth, R.K., Jackson, S.T., Thompson, T.A., 2002. Paleoecology of a northern Michigan
lake and the relationship among climate, vegetation, and Great Lakes water levels.
Quat. Res. 57, 120–130.
Booth, R.K., Jackson, S.T., Gray, C.E.D., 2004. Paleoecology and high-resolution
paleohydrology of a kettle peatland in upper Michigan. Quat. Res. 61, 1–13.
Bridge, J.S., 2003. Rivers and Floodplains. Blackwell, Oxford.
Bridgland, D., Maddy, D., Bates, M., 2004. River terrace sequences: templates for
Quaternary geochronology and marine-terrestrial correlation. J. Quat. Sci. 19,
203–218.
Bryan, R.B., Campbell, I.A., Fair, A., 1987. Postglacial geomorphic development of the
Dinosaur Provincial Park Badlands, Alberta. Can. J. Earth Sci. 24, 135–146.
Bull, W.B., 1990. Stream-terrace genesis: implications for soil development. Geomorphology 3, 351–367.
Bull, W.B., 1991. Geomorphic Responses to Climate Change. Oxford University Press,
New York.
Carson, M.A., 1984. The meandering-braided river threshold; a reappraisal. J. Hydrol. 73,
315–334.
Campbell, C., Campbell, I.A., 1997. Calibration, review, and geomorphic implications of
postglacial radiocarbon ages in southeastern Alberta, Canada. Quat. Res. 47, 37–44.
Chorley, R.J., 1963. Diastrophic background to twentieth-century geomorphological
thought. Geol. Soc. Amer. Bull. 74, 953–970.
Davis, W.M., 1909. Geographical Essays. Ginn and Company, Boston.
Davis, M., Winkler, M.G., Flakne, R., Douglas, C., Calcote, R., Cole, K.L., 2000. Holocene
climate in the western Great Lakes National Parks and Lakeshores: implications for
future climate change. Conserv. Biol. 14, 968–983.
Dury, G.H., 1970. Rivers and River Terraces. MacMillan, London.
Eichenlaub, V.L., Harman, J.R., Nurnberger, F.V., Stolle, H., 1990. The Climatic Atlas of
Michigan. University of Notre Dame Press, Notre Dame, IN.
Eschman, D.F., 1985. Summary of the Quaternary history of Michigan, Ohio, and Indiana.
J. Geol. Educ. 33, 167.
Evans, D.J.A., Campbell, I.A., Lemmen, D.S., 2004. Holocene alluvial chronology of One
Tree Creek, southern Alberta, Canada. Geogr. Ann. 86, 117–130.
Fisk, H.N., 1944. Geological Investigation of the Alluvial Valley of the Lower Mississippi
River. U.S. Army Corps of Engineers, Vicksburg. MS.
Harvey, A.M., Oldfield, F., Baron, A.F., Pearson, G.W., 1981. Dating of post-glacial
landforms in the central Howgills. Earth Surf. Processes Landf. 6, 401–412.
Howard, A.J., Macklin, M.G., Black, S., Hudson-Edward, K.A., 1999. Holocene river
development and environmental change in Upper Wharfdale, Yorkshire Dales,
England. J. Quat. Sci. 15, 239–252.
Howard, A.J., Macklin, M.G., Bailey, D.W., Mills, S., Andreescu, R., 2004. Late-glacial and
Holocene river development in the Teleorman Valley on the southern Romanian
Plain. J. Quat. Sci. 19, 271–280.
Hsieh, M.L., Knuepfer, P.L.K., 2001. Middle-late Holocene river terraces in the Erhjen
river Basin, southwestern Taiwan — implications of river response to climate
change and active tectonic uplift. Geomorphology 38, 337–372.
Jackson, S.T., Booth, R.K., 2002. The role of late Holocene climate variability in the
expansion of yellow birch in the western Great Lakes region. Divers. Distrib. 8,
275–284.
Kasse, K., Vandenberghe, J., Bohncke, S., 1995. Climatic change and fluvial dynamics of
the Maas during the late Weishselian and early Holocene. Palaoklimaforschung 14,
123–150.
Knighton, D., 1984. Fluvial Forms and Processes. Edward Arnold, New York.
Knox, J.C., 1972. Valley alleviation in southwestern Wisconsin. Ann. Assoc. Am. Geogr.
62, 401–410.
Knox, J.C., 1975. Concept of the graded stream. In: Melhorn, W.N., Flemal, R.C. (Eds.),
Theories of Landform Development. Publications in Geomorphology, State
University of New York, Binghamton, pp. 169–198.
Knox, J.C., 1984. Fluvial response to small scale climate changes. In: Costa, J.E.,
Fleisher, P.J. (Eds.), Developments and Applications of Geomorphology. Springer,
Berlin, pp. 318–342.
Knox, J.C., 1993. Large increases in flood magnitude in response to modest changes in
climate. Nature 361, 430–432.
Knox, J.C., 1995. Fluvial systems since 20,000 years BP. In: Gregory, K.J., Starkel, L., Baker,
V.R. (Eds.), Global Continental Palaeohydrology. John Wiley and Sons, Chichester,
pp. 87–108.
Knox, J., 2006. Floodplain sedimentation in the Upper Mississippi Valley: natural versus
human accelerated. Geomorphology 79, 286–310.
Kozarski, S., 1991. Warta — a case study of a lowland river. In: Starkel, L., Gregory, K.J.,
Thornes, J.B. (Eds.), Temperate Palaeohydrology: Fluvial Processes in the Temperate
Zone During the Last 15,000 Years. John Wiley and Sons, Chichester, UK, pp. 189–215.
Krishnamurthy, R.V., Syrup, K.A., Baskaran, M., Long, A., 1995. Late glacial climate record
of the Midwestern United States from the hydrogen isotope ratio of lake organic
matter. Science 269, 1565–1567.
Kutzbach, J., Gallimore, R., Harrison, S.P., Behling, P., Selin, R., Larrif, F., 1998. Climate and
biome simulations for the past 21,000 years. Quat. Sci. Rev. 17, 473–506.
Leigh, D.S., 2006. Terminal Pleistocene braided to meandering transition in rivers of the
southeastern USA. Catena 66, 155–160.
Leigh, D.S., Feeney, T.P., 1995. Paleochannels indicating wet climate and lack of response
to lower sea level, southeast Georgia. Geology 23, 687–690.
Linick, T.W., Suess, H.E., Becker, B., 1985. La Jolla measurements of radiocarbon in south
German oak tree-ring chronologies. Radiocarbon 27, 20–32.
Manny, B.A., Wetzel, R.G., Bailey, R.B., 1978. Paleolimnological sedimentation of organic
carbon, nitrogen, phosphorus, fossil pigments, pollen and diatoms in a hypereutrophic, hardwater lake: a case history of europhication. Pol. Arch. Hydrobiol. 25,
243–267.
Martin, C.W., 1992. The response of fluvial systems to climate change: an example from
the central Great Plains. Phys. Geogr. 13, 101–104.
May, D.W., 1989. Holocene alluvial fills in the South Loop valley, Nebraska. Quat. Res. 32,
117–120.
Minako, A., Tetsuji, M., 2006. The origin of stream terraces in the alluvial lower reaches
of rivers: an experimental examination of the complex response model. J. Geol. Soc.
Jpn. 112, 315–330.
Monoghan, G.W., Lovis, W.A., 2005. Modeling archaeological site burial in southern
Michigan: a geoarchaeological synthesis. Environmental Research Series No. 1.
Michigan Department of Transportation. Michigan State University Press, East
Lansing.
O'Neal, R.P., 1997. In: Division, F. (Ed.), Muskegon River watershed assessment. Michigan
Department of Natural Resources, Lansing.
Otelelaar, G.A., 2002. River of change: a model for the development of terraces along the
Bow River, Alberta. Geogr. Phys. Quat. 56, 155–169.
Penck, A., Bruckner, A.E., 1909. Die Alpen im Eiszeitalter. Tachnitz, Leipzig.
Pregitzer, K.S., Reed, D.D., Bornhorst, T.J., Foster, D.R., Mroz, G.D., McLachlan, J.S., Laks, P.E.,
Stokke, D.D., Martin, P.E., Brown, S.E., 2000. A buried spruce forest provides evidence at
the stand and landscape scale for the effects of environment on vegetation at the
Pleistocene/Holocene boundary. J. Ecol. 88, 45–53.
Rains, R.B., Welch, J., 1988. Out-of-phase Holocene terraces in part of the North
Saskatchewan River basin, Alberta. Canad. J. Earth Sci. 25, 454–464.
Schaetzl, R.J., Weisenborn, B., 2004. The Grayling Fingers region of Michigan: Sedimentology,
stratigraphy, and geomorphic development. Geomorphology 61, 251–274.
Schumanski, A., 1983. Palaeochannel of large meanders in the river valleys of the Polish
Lowland. Quat. Stud. Pol. 4, 207–216.
Schumm, S.A., Parker, R.S., 1973. Implications of complex response of drainage systems
for Quaternary alluvial stratigraphy. Nature 243, 99–100.
Schumm, S.A., Brakenridge, G.R., 1987. Stream responses. In: Ruddiman, W.F., Wright, H.E.
(Eds.), North America and Adjacent Oceans During the Last Deglaciation. The
Geological Society of America, Boulder, CO, pp. 221–240.
Scully, R.W., Arnold, R.W., 1981. Holocene alluvial stratigraphy in the upper
Susquehanna River basin, New York. Quat. Res. 15, 327–344.
Sidorchuk, A., Borisova, O., Panin, A., 2001. Fluvial response to the Late Valdai/Holocene
environmental change on the East European Plain. Glob. Planet. Change 28,
303–318.
Starkel, L., 1991. Long distance correlation of fluvial events in the temperate zone. In:
Starkel, L., Gregory, K.J., Thornes, J.B. (Eds.), Temperate Palaeohydrology: Fluvial
Processes in the Temperate Zone during the Last 15000 Years. Wiley, Chichester, UK,
pp. 473–495.
Starkel, L., 1995. Palaeohydrology of the temperate zone. In: Gregory, K.J., Starkel, L.,
Baker, V.R. (Eds.), Global Continental Palaeohydrology. Wiley, Chichester, UK,
pp. 233–258.
Starkel, L., Kalicki, T., Krapiec, M., Soja, R., Gebica, P., Czyzowska, E., 1996. Hydrological
changes of valley floors in upper Vistula basin during late Vistulian and Holocene.
In: Starkel, L. (Ed.), Evolution of the Vistula River Valley during the last 15000 years.
Geographic Studies, Special Issue 9, 7–128.
Straffin, E.C., Blum, M.D., Colls, A., Stokes, S., 2000. Alluvial stratigraphy of the Loire and
Arroux Rivers, Burgandy, France. Quaternaire 10, 271–282.
Stuiver, M., Polach, H.A., 1977. Discussion: reporting of 14C data. Radiocarbon 19, 355–363.
Stuiver, M., Pearson, G.W., Braziunas, T.F., 1986. Radiocarbon age calibration of marine
samples back to 9000 cal yr BP. Radiocarbon 28, 980–1021.
Stuiver, M., Reimer, P.J., Bard, E., Beck, J.W., Burr, G.S., Hughen, K.A., Kromer, B.,
McCormac, G., Van der Plicht, J., Spurk, M., 1998. INTCAL98 radiocarbon age
calibration, 24000–0 cal. B.P. Radiocarbon 40, 1041–1083.
Tardy, S.W., 2005. Soil Survey of Roscommon County, Michigan. U.S. Department of
Agriculture. Natural Resources Conservation Service, Washington, DC.
Taylor, R.E., 1987. Radiocarbon Dating: An Archaeological Perspective. Academic Press,
Orlando, FL.
Tebbens, I.A., Veldkamp, A., Westerhoff, W., Kroonenberg, S.B., 1999. Fluvial incision and
channel downcutting as a response to Late-glacial and Early Holocene climate
A.F. Arbogast et al. / Geomorphology 102 (2008) 615–623
change: the lower reach of the River Meuse (Maas), the Netherlands. J. Quat. Sci. 14,
59–75.
Thompson, T.A., 1992. Beach-ridge development and lake-level variation in southern
Lake Michigan. Sediment. Geol. 80, 305–318.
Thompson, T.A., Baedke, S.J., 1997. Strand-plain evidence for late Holocene lake-level
variations in Lake Michigan. Geol. Soc. Amer. Bull. 109, 666–682.
U.S. Geological Survey, 2004. Daily Water Data for Michigan. Retrieved January 31, 2004.
Van Nest, J., Bettis III, E.A., 1990. Postglacial response of a stream in central Iowa to
changes in climate and drainage basin factors. Quat. Res. 33, 73–85.
Walker, M.J.C., 1995. Climatic changes in Europe during the last glacial–interglacial
transition. Quat. Int. 28, 63–76.
Webb, I.T., Cushing, E.J., Wright, H.E., 1983. Holocene changes in the vegetation of the
Midwest. In: Wright, H.E. (Ed.), Late Quaternary Environments of the United States.
University of Minnesota, Minneapolis, MN, pp. 142–165.
623
Webb, I.T., Shuman, B., Williams, J.W., 2004. Climatically forced vegetation dynamics in
eastern North American during the Late Quaternary Period. In: Gillespie, A.R.,
Porter, S.C., Atwater, B.F. (Eds.), The Quaternary Period in the United States. Elsevier,
Amsterdam, pp. 459–478.
Woltemade, C.J., 1994. Form and process: fluvial geomorphology and flood-flow
interaction, Grant River, Wisconsin. Ann. Assoc. Am. Geogr. 84, 462–479.
Womack, W.R., Schumm, S.A., 1977. Terraces of Douglas Creek, Northwestern Colorado:
an example of episodic erosion. Geology 5, 72–76.